PeopleofSuperintelligence

Researcher

Eric Wallace

Eric Wallace is an AI researcher whose work spans language-model security, privacy, and alignment. He coauthored a 2018 EMNLP presentation on neural-model interpretation, then published a 2019 EMNLP paper on universal adversarial triggers for NLP and a 2021 USENIX Security paper on training-data extraction from language models. He received an Apple Scholars in AI/ML PhD fellowship in 2022 and pursued a Berkeley PhD with Dan Klein and Dawn Song. He coauthored the 2024 ICML Best Paper on model stealing, completed a Berkeley PhD on language-model vulnerabilities in 2025, and coauthored a study of worst-case risks from releasing gpt-oss. As of October 2026, he is an OpenAI researcher and co-leads Alignment Training.

Now at
OpenAI
Badges

Timeline

  1. 2018

    A neural-model paper

    In 2018, he coauthors “Pathologies of Neural Models Make Interpretation Difficult,” a research presentation at the Conference on Empirical Methods in Natural Language Processing.

  2. 2019

    Universal adversarial triggers

    In 2019, he coauthors “Universal Adversarial Triggers for Attacking and Analyzing NLP” at EMNLP.

  3. 2021

    Extracting LLM training data

    At USENIX Security in 2021, he coauthors “Extracting Training Data From Large Language Models.”

  4. 2022

    An Apple AI fellowship

    In 2022, he receives an Apple Scholars in AI/ML PhD fellowship.

  5. February 2023

    PhD research at Berkeley

    In February 2023, he is a fourth-year PhD student at UC Berkeley, advised by Dan Klein and Dawn Song. His research focuses on making language models more robust, trustworthy, secure, and private.

  6. 2024

    An ICML Best Paper

    He coauthors the ICML 2024 paper “Stealing part of a production language model,” which introduces an attack that extracts precise, nontrivial information from black-box production language models. The paper receives an ICML Best Paper Award.

  7. 2025

    Berkeley PhD and gpt-oss study

    In 2025, he completes a PhD in EECS at the University of California, Berkeley, with a thesis titled “Vulnerabilities of Language Models.” He coauthors “Estimating Worst-Case Frontier Risks of Open-Weight LLMs,” a study of worst-case risks from releasing gpt-oss.

  8. As of October 2026

    OpenAI's Alignment Training team

    As of October 2026, he is a researcher at OpenAI and co-leads Alignment Training, a team spanning safety, alignment, and capabilities.

Career

  1. NowOpenAIResearcher and co-lead, Alignment Training

Education

  1. 2025University of California, BerkeleyPhD in EECS

From the biography.

  1. 22 of 133 people shaping AI with a former OpenAI role now work at AnthropicIncludes Eric Wallace. Our estimates of where OpenAI recruits came from and where 133 people with former OpenAI roles work now, including the companies and labs they founded.
  2. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Eric Wallace. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  3. 16% of AI career moves in the 2020s went from a university to a companyIncludes Eric Wallace. We estimate that 16% of AI career moves in the 2020s went from a university to a company; 29% of moves starting at a university went to a company, and 76% ended the university role.
  4. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Eric Wallace. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Eric Wallace yet. Send us a free-licensed photo.

  1. New Method Peeks Inside the “Black Box” of Artificial Intelligence cmns.umd.edu. Checked 21 October 2026.
  2. Eric Wallace ericswallace.com. Checked 21 October 2026.
  3. Aviral Kumar, Serena Wang and Eric Wallace win 2022 Apple Scholars in AI/ML PhD fellowships eecs.berkeley.edu. Checked 21 October 2026.
  4. CLIP Colloquium: Eric Wallace (UC Berkeley) languagescience.umd.edu. Checked 21 October 2026.
  5. ICML 2024 Awards icml.cc. Checked 21 October 2026.
  6. Stealing part of a production language model proceedings.mlr.press. Checked 21 October 2026.
  7. People - The Berkeley NLP Group nlp.cs.berkeley.edu. Checked 21 October 2026.
  8. Vulnerabilities of Language Models www2.eecs.berkeley.edu. Checked 21 October 2026.
  9. Estimating Worst-Case Frontier Risks of Open-Weight LLMs arxiv.org. Checked 21 October 2026.